One of the major problems facing aquaculture in Ghana is the non-availability of quality and affordable fish feeds. The present study investigated the nutritional suitability and cost-effectiveness of some Ghanaian oilseed by-products, soybean meal (Glycine spp), cottonseed meal (Gossypium spp), groundnut cake (Arachis hypogaea L.) and groundnut husk, as alternative protein sources to fishmeal (FM) in the diet of Nile tilapia (Oreochromis niloticus L.). The oilseed meals were used individually, as mixtures, as mixtures enriched with methionine and mixtures detoxified by heat processing (autoclaving) and/or addition of supplements (viz. phytase and ferrous sulphate) intended to reduce levels of the most important antinutritional factors (ANFs). Diets, containing the oilseed meals at inclusion levels from 25% to 75% dietary protein, were formulated to be isonitrogenous (320 g.kg-1), isolipidic (100 g.kg-1) and isoenergetic (18 KJ.g-1) and fed to juvenile Nile tilapia at 4-10% of their body weight for a period of eight weeks. Proximate analysis showed that soybean meal (SBM), cottonseed meal (CSM), groundnut cake (GNC) and groundnut husk (GNH) had 500.3, 441.4, 430.5 and 205.6 g.kg-1 crude protein, 38.2, 89.5, 12.8 and 89.2 g.kg-1 crude fibre and 20.19, 19.61, 23.17 and 22.18 kJ.g-1 gross energy respectively. Generally the oilseed meals had good essential amino acid (EAA) profiles with the exception of GNH. The EAA profile of SBM compared very well with FM but methionine and threonine were low (0.73 and 1.50 % of protein respectively) and the same was true for CSM and GNC with even lower levels. Analyzed ANFs in SBM, CSM, GNC and GNH were 17.54, 31.64, 14.86 and 3.99 g.kg-1 phytic acid, 14.09, 1.24 and 2.34 g.kg-1 trypsin inhibitors and 5.80, 6.50, 8.01 and 10.08 g.kg-1 saponin respectively and in CSM 5.6 g.kg-1 gossypol. Nutrient digestibility of these oilseed proteins suggested that Nile tilapia may be able to utilize SBM, CSM and GNC efficiently as dietary protein sources due to high apparent protein digestibility of 94.50%, 84.93% and 90.01% respectively. However, GNH may not be suitable because of very low apparent protein digestibility (27.67%). These protein sources when used individually were shown to cause depressed growth and feed efficiency when substituting more than 50% of the FM protein in diets. This may be attributed to high levels of ANFs, high fibre content and poor EAA profile. However, the use of mixtures of these meals was found to be marginally more effective than that of single sources. This may have been as a result of lower levels of ANFs and improvement in essential amino acid profile due to mixing. Supplementing the mixtures with methionine led to improvement in feed utilization but without significantly improving the nutritive value compared with FM. Heat processing was effective in reducing heat labile trypsin inhibitors in SBM, CSM and GNC by almost 80%, but not phytic acid and saponins, which remained virtually unaffected. Use of meals detoxified by heat processing with/without supplements at 50% inclusion improved growth and feed utilization compared to the unprocessed meals and performance was generally not significantly different from FM. Cost effectiveness analysis revealed that diets containing single feedstuffs or mixtures, particularly those containing equal proportions of oilseed meals and higher proportion of CSM replacing between 50% - 75% FM protein, were more profitable than FM diet. Similarly, the use of heat processed meals at 50% replacement of FM protein yielded greater profit than all other diets including the FM diet. However, essential amino acid supplementation of the meals was less profitable compared to the control. Generally, fish fed diets with oilseed meals would take longer to attain harvest size compared with FM and this could lead to an increase in production costs or a decrease in the number of production cycles which could be achieved within a year. It can be concluded that there is nutritional and economic justification for using SBM, CSM and GNC as partial replacement for FM in diets of Nile tilapia. Based on growth performance, nutrient utilization and economic benefits the diet with heat processed oilseed meal mixtures (containing equal proportions of 16.67% each) at 50% inclusion has the best prospects for replacing FM protein in diets of O. niloticus.

Traditionally, fish meal (FM) and fish oil (FO) have been used extensively in aquafeeds, mainly due to their excellent nutritional properties. However, various reasons dictate the use of sustainable alternatives and the reduction of the dependence on these commodities in fish feeds. Hence, the aim of the present thesis was to investigate the effects of the replacement of FM and FO with two vegetable oils (VO) and an oilseed meal on the growth performance, feed utilization, nutrient and fatty acids (FA) digestibility and tissue FA composition and metabolism in three commercially important European fish species. Specifically, in Experiment I crude palm oil (PO) was used to replace FO in diets for rainbow trout. In Experiments II and III FO was replaced with rapeseed oil (RO) in diets for Atlantic salmon at various dietary protein/lipid levels aiming also at further reductions of FM by using low protein (high lipid) diet formulations. In Experiments II and III the fish were reared at low and high water temperatures, respectively, in order to elucidate, also, the potential effects of temperature. Lastly, the effects of the replacement of FM with full fat soya meal (FFS) in Atlantic cod were investigated in Experiment IV. The results of the present thesis showed no negative effects on growth performance and feed utilization in rainbow trout when FO was replaced with PO. The dietary inclusion of RO improved the growth of Atlantic salmon, possibly, due to changes in the nutrient and FA digestibilities and FA catabolism while, the growth and feed utilization were unaffected by the dietary protein/lipid level. However, the growth of Atlantic cod was affected negatively by the replacement of FM with FFS. The proximate composition of the fish whole body was in most cases unaffected by dietary treatments. The changes in dietary formulations affected the dietary FA compositions and resulted in significant changes in the fish tissue FA compositions. It was clearly shown that the fish tissue total lipid FA composition reflects the FA composition of the diet, although specific FA were selectively utilized or retained in the tissues by the fish. These may have serious implications not only for fish metabolism and growth but also for the quality of the final product, especially in terms of possible reductions of n-3 HUFA.